Sandia National Laboratories researchers combined large sets of real-world solar data and advanced machine learning to learn about the affects of extreme weather on U.S. solar farms, and type out what elements affect electricity generation. Their consequences have been posted in the past this month in the scientific journal Applied Energy.
Hurricanes, blizzards, hailstorms and wildfires all pose risks to solar farms both immediately in the form of costly damaged and not directly in the structure of blocked sunlight and reduced electricity output. Two Sandia researchers scoured protection tickets from more than 800 solar farms in 24 states and combined that statistics with electrical energy technology statistics and climate files to investigate the consequences of severe weather on the facilities. By identifying the factors that make a contribution to low performance, they hope to increase the resiliency of solar farms to extreme weather.
“Trying to understand how future climate conditions could impact our national energy infrastructure, is exactly what we need to be doing if we want our renewable energy sector to be resilient under a changing climate,” said Thushara Gunda, the senior researcher on the project. “Right now, we’re focused on extreme weather events, but eventually we’ll extend into chronic exposure events like consistent extreme heat.”
Hurricanes and snow and storms, oh my!
The Sandia lookup team first used natural-language processing, a kind of machine learning used by smart assistants, to analyze six years of p solar maintenance records for key weather-related words. The evaluation strategies they used for this find out about has considering been published and is freely available for other photovoltaic researchers and operators.
“Our first step was to look at the maintenance records to decide which weather events we should even look at,” said Gunda. “The photovoltaic community talks about hail a lot, but the data in the maintenance records tell a different story.”
While hailstorms tend to be very costly, they did not appear in solar farm maintenance records, likely because operators tend to document hail damage in the form of insurance claims, Gunda said. Instead, she observed that hurricanes had been referred to in nearly 15% of weather-related maintenance records, followed by the different weather terms, such as snow, storm, lightning and wind.
“Some hurricanes damage racking—the structure that holds up the panels—due to the high winds,” said Nicole Jackson, the lead author on the paper. “The other major issue we’ve seen from the maintenance records and talking with our industry partners is flooding blocking access to the site, which delays the process of turning the plant back on.”
Using machine learning to find the most important factors
Next, they combined more than two years of real world electricity production records from greater than 100 photo voltaic farms in sixteen states with historic climate statistics to assess the results of extreme weather on photo voltaic farms. They used records to discover that snowstorms had the highest effect on electrical energy production, observed via hurricanes and a established team of other storms.
Then they used a machine learning algorithm to discover the hidden elements that contributed to low performance from these extreme climate events.
“Statistics gives you part of the picture, but machine learning was really helpful in clarifying what are those most important variables,” said Jackson, who primarily conducted statistical analysis and the machine learning portion of the project. “Is it where the site is located? Is it how old the site is? Is it how many maintenance tickets were submitted on the day of the weather event? We ended up with a suite of variables and machine learning was used to home in on the most important ones.”
She observed that throughout the board, older solar farms have been affected the most by extreme weather. One possibility for this is that solar farms that had been in operation for more than five years had more wear-and-tear from being exposed to the elements longer, Jackson said.
Gunda agreed, adding, “This work highlights the importance of ongoing maintenance and further research to ensure photovoltaic plants continue to operate as intended.”
For snowstorms, which all at once had been the type of storm with the highest impact on electricity production, the next most important variables have been low sunlight tiers at the region due to cloud cover and the amount of snow, followed by several geographical features of the farm.
Expanding the approach to wildfires, the grid
As an unbiased research organization in this space, Sandia was once able to collaborate with multiple enterprise companions to make this work feasible. “We would no longer have been able to do this project without those partnerships,” Gunda said.
The research team is working to extend the project to find out about the effect of wildfires on photo voltaic farms. Since wildfires don’t seem to be mentioned in maintenance logs, they were not able to learn about them for this paper. Operators do not cease to write a maintenance file when their photo voltaic farm is being threatened by means of a wildfire, Gunda said. “This work highlights the reality of some of the data limitations we have to grapple with when studying extreme weather events.”
“The cool thing about this work is that we were able to develop a comprehensive approach of integrating and analyzing performance data, operations data and weather data,” Jackson said. “We’re extending the approach into wildfires to examine their performance impacts on solar energy generation in greater detail.”
The researchers are presently expanding this work to seem at the outcomes of severe weather on the entire electrical grid, add in greater manufacturing data, and answer even greater questions to assist the grid adapt to the altering climate and evolving technologies.
